Cosine Similarity

Category: science

A mathematical metric used to measure the structural closeness or correlation between two data vectors in a multi-dimensional space.

Cosine similarity calculates the cosine of the angle between two multi-dimensional embedding vectors ($Cosine\ \theta = (A \cdot B) / (||A||\ ||B||)$). In semantic search architectures, it measures conceptual alignment independent of text length: a value of 1.0 represents identical semantic meaning, while 0.0 indicates complete statistical orthogonality.

Common Examples

• Our ClickHouse database utilizes vectorized cosine similarity matching to pull articles that share an identical bias profile with the seed document.
• Calculating cosine similarity across millions of high-dimension rows requires hardware-accelerated instruction sets like AVX-512 to avoid latency spikes.